The mechanisms of 39UTR-dependent nanos mRNA regulation have been addressed previously in fishes and flies, in which miR430 and the Dnd1 protein, or the Smg, Glo and Osk proteins are involved in mRNA regulation via the nanos-39UTR. In mice, one ortholog of Dnd1 and two orthologs of Smg have now been identified [40,41]. We examined the possible effects of these proteins on the translation of an mRNA harboring theNanos3- 39UTR by a luciferase assay inthe NIH3T3 cell line. The stability of luciferase mRNA is also affected by Nanos3-3 9UTR. However, the addition of both proteins did not result in any effects on reporter activities (data not shown). It is possible that they need co- factors which are not expressed in this cell line. It is also possible that the abundant expression of endogenous Smg in NIH3T3 cells caused no effect. In addition, the sequence of Nanos3-39UTR has almost no similarity to the 39UTRs of nanos orthologs and has no significant match with any miRNA target sites. Although several stem-loop structures have been predicted using the ‘mfold’ program (Zuker, 2003), these are not similar to the Drosophila TCE (data not shown). Hence, the mechanism of Nanos3-3 9UTR dependent regulation is still unclear and is an essential project for a future study.
Knocking down Sirt2 expression does not affect the self-renewal of mouse ESCs (Fig. 2), however during EB differentiation, the absence of Sirt2 inhibits the formation of ectoderm while promoting the differentiation of mesoderm and endoderm; this Figure 3. Activated GSK3b is involved in Sirt2 stable knockdown ESC differentiation. (A) Western blots analyzing pGSK3b and total GSK3b levels in control (shpLKO.1) and Sirt2 knockdown cells (shSirt2-1) using GAPDH as a loading control. (B) The relative protein levels of pGSK3b and total GSK3b levels in GSK3b mutant cell lines. (C) AP staining of control and constitutively active GSK3b mutant S9A cell lines. All Figures 1006. (D) Real- time PCR analysis for the self-renewal marker genes Oct4, Nanog and Rex1. (E) Marker genes of all three germ layers in EBs generated from control and GSK3b-S9A cells at day 6 are verified by real-time PCR. Data were normalized to Gapdh mRNA expression levels and are means 6SEM (n = 3). P,0.05(*), P,0.01(**) and P,0.001(***) vs. control EBs generated from FUGW-infected mouse ESCs (Student’s t-test). (F) Immunofluorescence staining for Tuj1, Actin and Gata4 in EBs generated from control and GSK3b-S9A ESCs at day 9. Cells were counterstained with DAPI (blue). All Figures 1006.
Inthe present study, we generated Pik3c3 knockout mice and characterized thein vivo function of PIK3C3 during embryogenesis. The most prominent defect caused by Pik3c3 deletion is a severely reduced embryonic cell proliferation. The death of Pik3c3 mutant embryos prior to gastrulation is likely due to the lack of proliferative burst required for the development of different germ layers. It has been shown that mouse gastrulation occurs at around E6.5 when mesoderm is generated from epiblast . Importantly, there is normally a ,100-fold increase incell number between E5.5 and E7.5 . The dramatic increase incell proliferation is necessary for proper gastrulation since cells need to accumulate to certain numbers to initiate this process . In fact, mutations in genes that affect proliferation rate during this period often result in missing of mesoderm, abnormal embryonic development and early embryonic lethality . For instance, embryos carrying null mutation in genes encoding proteins of the bone morphogenetic protein (BMP) signaling pathway, including Smad4 [50,51], Bmp4  and Bmpr , all fail to develop mesoderm and die at the beginning of gastrulation partly due to defects incell proliferation. Germline deletion of tumor suppressor genes regulating cellular growth such as Brca1  and Tsg101  also leads to cell proliferation defect, lack of mesoderm and early embryonic lethality before E7.5. Here we
To inhibit Geminin expression we injected two-cell Xenopus embryos with 32 ng of anti-Geminin morpholino oligonucleotides (a-Gem MOs), which cause a .95% reduction inthe amount of Geminin protein through at least the gastrula stage (Fig. 1A, ). When theembryos reached the onset of gastrulation (Nieuwkoop stage 10.5) we compared the levels of a broad array of early transcripts in Geminin-depleted and control embryos by quanti- tative real-time PCR (RT-PCR). Geminin-depleted embryos showed a marked reduction intheexpression of most early transcripts (Fig. 1B), including patterning genes of the Spemann organizer (Goosecoid (Gsc)), genes inthe Wnt signaling pathway (Xwnt8, Sizzled (Szl)), genes inthe BMP signaling pathway (Chordin (Chrd), Vent2), and genes expressed inthe three germ layers (Brachury (Xbra, mesoderm), epidermal keratin (Epiker, epidermis), Zic3 (neural plate), and Sox17 (endoderm). The transcript levels of two genes expressed before the MBT, Nodal- related 5 (Xnr5)  and ID2 , were not significantly affected. Only two genes, Msx1b and Xenopus-posterior (Xpo), showed preserved induction in Geminin-depleted embryos. These results show that Geminin depletion causes widespread defects in zygotic transcription but has less of an effect on genes expressed before the MBT. The decrease in gene expressionis not specific to any particular signaling pathway, organ system, or germ layer.
Several protocols for inducing germcell differentiation from ES cells have been reported. In mice, germ cells have been generated from ES cells using monolayer culture , the formation of embryoid bodies (EBs) [2,3], co-aggregation with BMP4-produc- ing cells , and the use of mouse testicular cell-conditioned medium . In humans, germcell differentiation from ES cells via spontaneous EB formation, and EB formation with recombinant human bone morphogenetic proteins (BMPs) has been reported [5,6]. In monkeys, methods for inducing germcell differentiation from ES cells have not been reported except spontaneous germcell differentiation by EB formation . Therefore, it is very important to develop a suitable protocol to induce in vitro germcell differentiation from monkey ES cells before non-human primate ES cells can be used as a model for in vitro differentiated germ cells. The current study examined theexpression of germcell marker genes in tissues and ES cells of the cynomolgus monkey, and theexpression of several germcell marker genes including VASA was confirmed. The up-regulation of VASA expression was observed in ES cells differentiated via spontaneous EB formation. Theexpression of other germcell marker genes, such as NANOS1, NANOS2, NANOS3 and PIWIL1, increased inthe EBs as well. SSEA1-, OCT-4-, and VASA-positive cells were detected inthe EBs, indicating that monkey ES cells have the ability to differentiate into a germcell lineage in vitro. In addition, the effects of mouse testicular  and ovarian cell-conditioned media, and also BMP4, RA, and SCF, on germcell differentiation in monkeys were also investigated.
Preimplantation development inthemouseis a time of dynamic change in which the fertilized egg becomes a pluripotent embryo that subsequently develops into a blastocyst with two distinct cell lineages. This developmental period is characterized by three major transitions, each of which entails pronounced changes inthe pattern of gene expression. The first transition isthe maternal-to- zygotic transition (MZT) that serves three functions: (1) to destroy oocyte-specific transcripts, (2) to replace maternal transcripts that are common to the oocyte and early embryo with zygotic transcripts and (3) to facilitate the reprogramming of the early embryo by generating novel transcripts that are not expressed inthe oocyte . Zygotic gene activation initiates during the late 1- cell stage at some genes and throughout the genome by the 2-cell stage [1,2]. Coincident with genome activation isthe acquisition of a chromatin-based transcriptionally-repressive state [3,4] and more efficient use of TATA-less promoters , which are likely to play a major role in establishing the appropriate patterns of gene expressionrequired for proper development.
Setdb1, also known as Eset and KMT1E, is a lysine methyltransferase (KMT) specific for the repressive histone H3 lysine 9 di- and tri-methyl (H3K9me2/me3) marks [14,15]. It is associated with transcriptional repression of euchromatic genes and maintenance of hetero- chromatin structure [14,15,16]. Recent evidence suggests that Setdb1 also plays a critical role in silencing retrotransposons in undifferentiated embryonic stem (ES) cells, as well as in early embryos and primordial germ cells (PGCs), where DNA methylation levels are low due to epi- genetic reprogramming [17,18]. DNA methylation isrequired for retrotransposon silencing in somatic cells . Setdb1 is an evolutionally conserved gene. Its Drosophila ortholog dSetdb1 (also known as dEset and Eggless) is involved in multiple developmental processes, including oogenesis [20,21,22]. Mouseembryos lacking Setdb1 die at the peri-implantation stage (around 3.5–5.5 days post coitum (dpc)) , which is significantly earlier than the phe- notypes of mice deficient for other H3K9 KMTs, such as Suv39h1/Suv39h2 (developmental defects after ~12.5 dpc)  and G9a (lethality at ~9.5 dpc) . Setdb1 is present at high lev- els in oocytes and zygotes and persists through preimplantation development [26,27]. How- ever, expression of zygotic Setdb1 is undetectable until the blastocyst stage [23,26]. These observations suggest that maternal Setdb1 may play important roles in oogenesis and/or early embryogenesis.
lization of target mRNAs . While this manuscript was under revision, Dicer-dependent but Drosha-independent endo-siRNAs were reported in male germ line that can function in post- transcriptional control of a wide variety of protein encoding mRNAs . Therefore, it is likely that at least some of the defects in Dicer1 knockout testes are due to the absence of miRNA and/or endo-siRNA-mediated mRNA regulation. Post-transcriptional control is of central importance during haploid differentiation due to the transcriptional silencing that results from the tight packing of chromatin with protamines [2,4]. Since the most striking defects inthe differentiation of Dicer-null male germ cells coincide chronologically with chromatin condensation and transcriptional silencing, it can be envisaged that Dicer is involved in this post-transcriptional control of haploid mRNAs. High throughput transcriptome analysis as well as detailed quantitative proteomics will uncover the genes that are under the regulation of Dicer-generated small RNAs during male germcell differentiation. piRNAs have a well-characterized role in transposon silencing in male germ cells , and several studies have also linked Dicer- dependent pathways with the regulation of transposon expression [14,15,18,34]. The most extensive transposon derepression in male germ cells takes place during epigenetic reprogramming in PGCs and in very early postnatal cells (from embryonic day 15 to 3 dpp), which enables the establishment of novel sex-specific epigenetic marks inthe genome . The level of transposable element transcripts seemed unchanged in our Dicer1 knockout model (Fig. 7). In contrast, transposon expression was increased inthe spermatocytes of Dcr(fx/fx); Ddx4Cre mice, which shows efficient Dicer1 recombination already at birth (Romero et al., co-submitted manuscript). This difference may be explained by the later Cre expressionin Dcr(fx/fx); Ngn3Cre testes only after major epigenetic
Normal pole cell migration requires the protein products of several genes such as serpent (srp) and huckebein (hkb), engaged in normal mesoderm development [27,28,29] and others that affect pole cell migration per se, such as wunen (wun), necessary for pole cell migration along the basal surface of the midgut . In addition, mutations in Abdominal A (abdA), Abdominal B (abdB), tinman (tin), heartless (htl), fear of intimacy (foi), trithorax (trx), trithoraxgleich (trg) kai zinc-finger homeodomain-1 (zfh-1) affect normal pole cell migration and result in defective gonad formation . Interestingly, mutations in many of these genes precipitate changes highly reminiscent of the phenotype of D14-3-3e homozygotes, namely scattered pole cells across the mesoderm and gonads devoid of, or containing few germ cells. A potential explanation for the phenotypic similarity in these loss-of-function mutants is that interaction among one or more of these proteins and D14-3-3e is necessary for pole cell migration and/or gonad formation. In fact, an in silico search revealed that Abdominal A, Columbus, Trithorax and Zinc-finger homeodomain-1 proteins contain one or more potential 14-3-3 binding sites (Table S2). Hence, we investigated whether the distribution pattern of these proteins was altered in D14-3-3e mutant embryos. We focused on Zfh-1 for two reasons. It isthe only one of the group that contains a perfect match to the consensus of the most common 14-3-3 binding site, Arg-Ser-x-Ser-x-Pro (where a x is any aminoacid) and has a highly specific antibody  available.
Nanos was irst identiied in Drosophila, where it represses the translation of target mRNAs through binding to their 3 � UTR and has a conserved function ingermcell development across species. Members of the evolutionarily conserved Nanos gene family are preferentially expressed inthe ovaries and are known to play an important role ingermcell devel- opment, maintenance, and survival [24–30]. In Drosophila, the single Nanos gene (Nos) isrequired for development of the abdomen as well as for germ line maintenance [31, 32]. hree Nanos homologues exist inmouse, with Nanos2 and Nanos3 functioning primarily in male germcell development and maintaining PGCs viability, respectively [33, 34]. In mice, Nanos3is expressed inthe primordial germ cells (PGCs) from their formation until shortly ater their appearance inthe gonads (E13.5 in female and E14.5 in male embryos) . Male and female mice deicient inNanos3 are infertile, and female �����3 −/− mice have atrophic ovaries in which no germ cells are detectable due to loss of migrating PGCs during embryogenesis . PGCs are lost by apoptosis inthe absence of Nanos3, establishing an essential function of Nanos3 as a repressor of apoptosis inthegermcell .
Relating expression signatures from different sources such as cell lines, in vitro cultures from primary cells and biopsy material is an important task in drug development and translational medicine as well as for tracking of cell fate and disease progression. Especially the comparison of large scale gene expression changes to tissue or cell type specific signatures is of high interest for the tracking of cell fate in (trans-) differentiation experiments and for cancer research, which increasingly focuses on shared processes and the involvement of the microenvironment. These signature relation approaches require robust statistical methods to account for the high biological heterogeneity in clinical data and must cope with small sample sizes in lab experiments and common patterns of co-expressionin ubiquitous cellular processes. We describe a novel method, called PhysioSpace, to position dynamics of time series data derived from cellular differentiation and disease progression in a genome-wide expression space. The PhysioSpace is defined by a compendium of publicly available gene expression signatures representing a large set of biological phenotypes. The mapping of gene expression changes onto the PhysioSpace leads to a robust ranking of physiologically relevant signatures, as rigorously evaluated via sample-label permutations. A spherical transformation of the data improves the performance, leading to stable results even in case of small sample sizes. Using PhysioSpace with clinical cancer datasets reveals that such data exhibits large heterogeneity inthe number of significant signature associations. This behavior was closely associated with the classification endpoint and cancer type under consideration, indicating shared biological functionalities in disease associated processes. Even though the time series data of cell line differentiation exhibited responses in larger clusters covering several biologically related patterns, top scoring patterns were highly consistent with a priory known biological information and separated from the rest of response patterns.
Trichostatin A (TSA) is a histone deacetylase inhibitor that induces histone hyperacetylation and increases gene expression levels. The aim of the present study was to establish a suitable condition for the use of TSA inin vitro cultures of bovine embryos, and to determine whether TSA would increase blastocyst rates by improvement of chromatin remodelling during embryonic genome activation and by increasing theexpression of crucial genes during early development. To test this hypothesis, 8-cellembryos were exposed to four concentrations of TSA for different periods of time to establish adequate protocols. In a second experiment, three experimental groups were selected for the evaluation of embryo quality based on the following parameters: apoptosis, total cell number and blastocyst hatching. TSA promoted embryonic arrest and degeneration at concentrations of 15, 25 and 50 nM. All treated groups presented lower blastocyst rates. Exposure of embryos to 5 nM for 144 h and to 15 nM for 48 h decreased blastocyst hatching. However, the terminal deoxynucleotidyl transferase-mediated dUTP nick-end labeling assay (TUNEL) assay revealed similar apoptosis rates and total cell numbers in all groups studied. Although, inthe present study, TSA treatment did not improve the parameters studied, the results provided background information on TSA supplementation during in vitro culture of bovine embryos and showed that embryo quality was apparently not affected, despite a decrease in blastocyst rate after exposure to TSA.
The observed cell dissociation in response to excess March8 focused our attention on cadherins, major cell-cell adhesion proteins that have been well studied in Xenopus and zebrafish embryos [3,14,38,39]. Xenopus [9,10,11,12], and zebrafish [4,5,6,7] cadherins are expressed at high levels inthe egg and early embryo, and are essential for cleavage, epiboly and gastrulation movements [8,40,41]. The essential function of E- cadherin in zebrafish epiboly is most clearly demonstrated inthe half baked mutant, which affects the cdh1 locus . Regulation of cadherin function is thus necessary for development, but due to the maternal expression of E-cadherin such regulation is expected to involve posttranscriptional, and likely posttranslational, mecha- nisms. It has been shown that E-cadherin activity can be regulated by Galpha12/13 binding to its intracellular domain, so that changes in Galpha12/13 action lead to cell dissociation and abnormal cell movements . Further, appropriate intracellular localization of E-cadherin is essential: mutant embryos lacking transcription factor Pou5f1 show delayed gastrulation movements due to misregulation of E-cadherin endosomal trafficking, mediated by EGF regulation of p120 . E-cadherin is a substrate for ubiquitination by the RING class E3 ubiquitin ligase Hakai, which leads to its endocytosis in epithelial cells . The possible function of Hakai in vertebrate embryosis unknown, but the Drosophila homolog isrequired for normal development inthe fly . We found that March8 can lead to removal of E-cadherin from thecell surface, both inembryos and in cultured cells. Inthe context summarized above these findings suggest that regulating E-cadherin localization and abundance inthe embryo is a biological role of March8, thereby making a contribution to the appropriate balance between cell-cell adhesion required for stability and relaxation of adhesion required for cell motility, both of which are indispensible for development.
The occurrence of balancing selection at the HLA-G promoter region has previously been described (Tan et al., 2005). Considering that HLA-G expression and levels were already related to different situations (either in physiologi- cal or pathological conditions) and that this molecule, depending on the context, may be deleterious or advanta- geous, balancing selection at this locus seems to be a plausi- ble possibility. Recently, Castelli et al. (2011) made a comprehensive review of the HLA-G gene polymorphism and haplotypes in a Brazilian urban cohort, evidencing a high linkage disequilibrium along the whole length of the gene. In this same work, the authors revealed evidence for balancing selection acting on the regulatory regions only (5’ and 3’ UTRs) and on the HLA-G locus as a whole. We cannot rule out that the evidence of balancing selection ob- served in our data could be due to a hitchhiking effect caused by a linkage disequilibrium between the 14 bp locus and the HLA-G promoter region. Nevertheless, the compel- ling evidence for the functionality of the 14 bp insertion in alternative splicing and its potential role in post-transcrip- tional regulation by microRNA binding make us believe that the 14 bp INDEL might also be an adaptive factor, in- fluencing HLA-G expression patterns and probably is re- lated to survival of heterozygous fetuses due to resistance to pathogens (Mendes-Junior et al., 2007). In conclusion,
Figure 2. Single embryo gene expression analysis. Total cDNA was PCR amplified from poly-A+ RNA prepared from individual wildtype or mutant 12-cell stage embryos (indicated above each blot compilation), separated briefly on an agarose gel side-by-side with like-staged samples of the same genotype, and blotted to membrane (’pseudo Northern’ blot; see Materials and Methods). Replicate blots were hybridized with 32 -P labeled probes prepared from the end-1, end-3, and sdz-23 genes, and the tba-1 (alpha tubulin) gene as a loading control. Exposures were to x-ray film which was then scanned. All end-1, end-3 and sdz-23 exposures were for a similar length of time (18–20 hours), whereas the tba-1 exposures were much shorter (10–15 minutes). itDf2 is a deletion that removes the genomic region containing both end-1 and end-3 (along with a number of other genes). Dots below lane numbers denote samples from individual mom-1(or10), mom-2(or42), mom-3(or78) and mom-4(ne19) mutant embryos that express greatly reduced combined levels of end-1 and end-3.
agreement with the overall reduction in courtship (Table 1). These data show that males with feminized blood–brain barrier are capable of all steps of courtship, but perform them with reduced probability. This is not due to locomotion defects, since males with feminized bbb perform indistinguishably from control flies in a short term activity assay  (Figure 1g). In this assay, individual males are placed in a courtship chamber and the number of time they cross a drawn line is counted. To corroborate that the observed courtship reduction is caused by the feminization of glial cells, we used repo-Gal4, a driver that is expressed not just inthe bbb but generally in glial cells . We observed a similar reduction, confirming that the male identity of glial cells is important for male courtship (Figure 1b, 1h). This would predict that a similar effect should be observed if these cells were made ‘‘less male’’. Since TraF acts through its downstream targets fru and dsx we next examined the effect of expressing RNAis that target these transcripts in bbb cells. Indeed, SPG-Gal4/UAS- fruRNAi and SPG-Gal4/UAS-dsxRNAi males showed similar reduc- tions as SPG-Gal4/UAS-TraF animals (Figure 1c, 1i). This suggests that FRU and DSXM both have a role in regulating sex-specific molecules inthe bbb. It also argues that the effect of TraF is not due to merely overexpressing female-specific DSXF.
To investigate the mechanism of Gas2-induced cell division arrest and cytokinesis failure, we used a wound-induced contractile array assay in Xenopus oocytes, which is also a model of cytokinesis . This assay permitted us to study Gas2-cytoskeletal interac- tions during microtubule-dependent actomyosin array formation and contraction in vivo . Oocyte wound healing and cytokinesis are highly dynamic processes that require coordination between the actin and microtubule cytoskeletons . We have shown that Gas2 possesses both actin- and tubulin-binding properties in vitro; therefore, we propose that Gas2 functions as a structural cross- linking protein between these two cytoskeleton systems. When oocytes are treated with taxol to stabilize microtubules prior to wounding, an abnormal double ring of actin forms around the wound border . Staining with phalloidin for F-actin showed that the internal ring is composed of contractile F-actin, while staining for actin monomers showed that the outer ring is formed by de novo actin polymerization [2,13]. Interestingly, we found that when full-length Gas2 or its Gas2 domain was over-expressed, oocytes developed the abnormal double actin ring phenotype upon wounding. Thus, Gas2 mimics the taxol-treatment phenotype, suggesting that Gas2 has microtubule-stabilizing activity. The Gas2-induced double rings can be rescued by de-polymerizing microtubules with nocodazole (Fig. 4L–O) perhaps because the Gas2 protein binds dynamically to microtubules. Full-length Gas2 co-localizes with the actin rings, presumably by binding actin via its CH domain (Fig. 4H–K). It is interesting to note that the Gas2 domain alone also forms a ring structure and localizes between two actin rings during the wound healing (Fig. 4T–W), but it does not overlap with either actin ring since it has no actin-binding domain.
We investigated two cases in which B. aphidicola have been lost during the evolution of aphids. Given our observations that bacteria are not required for the developmental maintenance of bacteriocytes, it is possible that the bacter- iocyte cell type might be lost if it had no other function. This does not appear to be the case. Inthe lineage including T. styraci, B. aphidicola was lost and a eukaryotic ‘‘yeast-like’’ symbiont has been gained (Buchner 1965; Fukatsu and Ishikawa 1992a; Fukatsu et al. 1994). Buchner (1965) suggested that the bacteriocytes of Cerataphis freycinetiae, another species inthe same lineage, are originally speciﬁed, become polyploid and then degenerate. We found Dll-expressing putative bacteriocyte nuclei to be speciﬁed and maintained over extensive periods of embryonic development in T. styraci. Buchner documented considerable variation inthe details of symbiotic transmission and bacteriocyte development, and it is possible that bacteriocyte development proceeds along different paths in these two species.
BRM has a crucial role in vegetative, embryonic and reproductive plant development [5,8,11,12]. Expression profiling using 10-day-old brm and wild-type (WT) seedlings showed that only 1% of the genes were differentially expressed in brm . However, when the same experiments were carried out with leaves from 14-day-old seedlings, the number of misregulated genes was more than 4% . These different results could indicate tissue and stage specificity for BRM-mediated gene expression. BRM is also required for the floral transition. Four main genetic pathways have been described that control flowering in Arabidopsis: the photoperiod pathway (day lengths), the vernalization pathway (prolonged cold temperature experienced during winter), the gibberellin pathway (gibberellins) and the autonomous pathway (repression of FLC) [15,16]. These different routes converge at the regulation of the integrator genes that play a crucial role inthe regulation of floral transition. Transgenic plants with reduced expression of BRM (BRM-silenced plants) showed an early- flowering phenotype in long day and short day conditions (LD and SD respectively) and these results were correlated with an increase intheexpression of the flowering integrator gene FLOWERING LOCUS T (FT) and the photoperiod-pathway gene CONSTANS (CO) . brm mutants showed a most dramatic phenotype than BRM-silenced plants with a slow growth, delayed development and a strong plant size reduction. The brm mutants flowered with less leaves than WT plants, but a percentage of the mutant plants never flowered under SD . These data indicate a more complex scenario for the involvement of BRM in flowering, which prompted us to carry out an in depth characterization. We show here that BRM is not only involved in regulation of the photoperiod pathway genes, but it is also an essential repressor of FLOWERING LOCUS C (FLC).
multiple duplication events . The respective coding regions and exon/intron boundaries are highly conserved among orthologues and paralogues, which is reflected by the redundant functions that Eph/ephrins can fulfill and by the ability of Eph receptors to interact with multiple ephrins. Moreover, the3’UTRs of Eph/ephrin orthologues are highly conserved between human and mouse (up to 80%), which indicates strong selection pressure and perhaps conservation of distinct expres- sion profiles. In contrast, the3’UTRs of Eph/ephrin paralogues show a high degree of structural divergence, which may have enabled Ephs/ephrins to acquire diverse expression profiles (cellular or subcellular), as has been suggested for other conserved paralogues . Despite their poor overall conser- vation, the3’UTRs of 7 of 8 ephrin ligands and 11 of 13 Eph receptors contain CPEs, AREs, and/or HuR binding sites. The grouping of several of these motifs into evolutionarily conserved clusters further emphasizes their functional importance. How- ever, because neither the position nor composition nor distribution of these motifs is conserved between paralogous Eph/ephrins, the clusters may have arisen independently in evolution. Such convergent evolution has been described for regulatory elements of other genes coding for proteins that functionally interact [27,28,29,30]. This isin line with our observation that conserved motif clusters are present in functionally interacting Eph/ephrins. For example, transcripts of the EphA2 receptor and its ligand, ephrin-A1, both contain conserved clustered motifs, as do the EphA4 receptor and its ligands, ephrin-A1 and ephrin-B2. Several clustered motifs are located close to poly(A) sites, which may reflect a function for these clusters in polyadenylation, as discussed below.